Arch. Min. Sci., Vol. 59 (2014), No 4, p.

9871004
Electronic version (in color) of this paper is available: http://mining.archives.pl

DOI 10.2478/amsc-2014-0068

UKASZ KLIMKOWSKI*, STANISAW NAGY*

KEY FACTORS IN SHALE GAS MODELING AND SIMULATION

MODELOWANIE I SYMULACJA NIEKONWENCJONALNYCH Z

GAZU ZIEMNEGO Z UPKW - KLUCZOWE PARAMETRY

Multi-stage hydraulic fracturing is the method for unlocking shale gas resources and maximizing ho-
rizontal well performance. Modeling the effects of stimulation and fluid flow in a medium with extremely
low permeability is significantly different from modeling conventional deposits. Due to the complexity
of the subject, a significant number of parameters can affect the production performance. For a better
understanding of the specifics of unconventional resources it is necessary to determine the effect of various
parameters on the gas production process and identification of parameters of major importance. As a result,
it may help in designing more effective way to provide gas resources from shale rocks.
Within the framework of this study a sensitivity analysis of the numerical model of shale gas reservoir,
built based on the latest solutions used in industrial reservoir simulators, was performed. The impact of
different reservoir and hydraulic fractures parameters on a horizontal shale gas well production perfor-
mance was assessed and key factors were determined.

Keywords: shale gas, unconventional gas, reservoir simulation, hydraulic fractures, shale gas extraction,
stimulation, hydraulic fracturing, key parameters, sweet spot, Blue Gas.

W celu udostpnienia zasobw gazu ziemnego ze z upkowych i maksymalizacji wydajnoci ho-

ryzontalnych odwiertw eksploatacyjnych stosowane jest wielostopniowe szczelinowanie hydrauliczne.
Modelowanie efektw stymulacji oraz przepywu pynw w orodku o ekstremalnie niskiej przepuszczal-
noci jakim jest skaa upkowa rni si znacznie od modelowania z konwencjonalnych. W zwizku
ze zoonoci zagadnienia wystpuje znaczna ilo parametrw majcych wpyw na przebieg wydoby-
cia. Dla lepszego zrozumienia specyfiki z niekonwencjonalnych konieczne jest okrelenie wpywu
poszczeglnych parametrw na cao procesu wydobycia gazu oraz identyfikacja tych o najwikszym
znaczeniu. W efekcie moe si to przeoy na projektowanie bardziej efektywnego sposobu udostpnienia
zasobw gazu ze z upkowych.
W ramach niniejszego opracowania przeprowadzono analiz wraliwoci numerycznego modelu
symulacyjnego zoa gazu z upkw zbudowanego w oparciu o najnowsze rozwizania stosowane w prze-

* AGH UNIVERSITY OF SCIENCE AND TECHNOLOGY, FACULTY OF DRILLING, OIL AND GAS, AL. A. MICKIEWICZA 30,
30-059 KRAKOW, POLAND

Unauthenticated
Download Date | 8/7/17 10:24 AM
988

mysowych symulatorach zoowych. W wyniku tej analizy okrelono parametry o kluczowym wpywie
na przebieg eksploatacji i maksymalizacj wydobycia gazu ze zoa.

Sowa kluczowe: gaz z upkw, gaz niekonwencjonalny, symulacja numeryczna, szczeliny hydrauliczne,
wydobycie gazu, stymulacja, szczelinowanie hydrauliczne, kluczowe parametry, sweet
spot, Blue Gas

1. Introduction
Extraction of natural gas from shale rocks, tight rocks (also production of coal bed meth-
ane) successfully implemented in the USA in the last decade is a game changer technology.
This process stimulates the American economy (IEA, (2011, 2012); EIA 2013). Natural gas is
the most important energy carrier with high public acceptance and according to the IEA reports
world gas demand will increase in the next 30 years (IEA (2011, 2012)). This increase of natural
gas demand in power generation processes also results in the development of gas combined
cycle technology (CCGT), reaching a high cycle efficiency more than 50% (Siemek & Nagy,
2012). Natural gas is also an important energy source for the transition time towards a low-carbon
economy (IEA 2011, 2012; MIT 2011; Siemek et al., 2009; Nagy & Siemek, 2011). The high gas
import dependency (Siemek et al., 2019; Nagy et al., 2009) leads Poland to search own way to
increase its energy safety in next twenty years.
As the natural gas (and/or oil) recovered from the shale rocks is quite conventional, the
term unconventional refers to methods, knowledge and approach applied to recover hydro-
carbons from such type of reservoirs (in comparison with so called conventional reservoirs)
rather than to natural gas itself.
Massive hydraulic fracturing treatment, till now mainly hydraulic, is the only method that
allows for economically effective gas production rates from unconventional shale gas reservoirs.
However, one can say that unconventionality of these reservoirs decreases with time as the
knowledge and technology increases and that nowadays shale gas is bit less unconventional
than it was few years ago.
Similar situation is observed in terms of modeling and simulation of unconventional res-
ervoirs. The technology of shale gas production in a certain sense surprised, or ahead of, the
creators of numerical reservoir simulators. Models used earlier did not take into account (and
to some extent so far) the specificity of shale rock reservoirs. Modeling deposits of this type
therefore requires methods at least equally unconventional, as the way to make them available.
Recently it is becoming more and more obvious that the Polish gas-bearing shales differ
significantly from US deposits. Unfortunately (after the Polish experience) the technology should
be modified to specific geological and local requirements and the new technology has to apply
with modification to compensate the variety of properties of reservoir rocks (Poprawa, 2010;
PGI, 2012; Nagy et al., 2013; Siemek et al., 2013; Kaliski et al., 2012).
The differences are so significant that there is no obvious deposits of analogous to which
we can refer and directly benefit from the experience of US companies. In view of the differences
in the parameters describing the so-called shale reservoirs the need of probably other than the
American (in some extent) methods of stimulation and development arises. Moreover, also there
is a need to understand the mechanisms of flow in the Polish shale for their proper modeling
and simulation of production of hydrocarbons (gas and/or oil/gas condensate). The lack of gas
production profile of polish shale well with ongoing continuous production strongly hinders the

Unauthenticated
Download Date | 8/7/17 10:24 AM
 989

issue. However, it seems that one should look for a way to understand the relevant mechanisms
of the gas flow in shale rocks, to model it effectively and consequently provide an effective way
of development of unconventional hydrocarbon resources in Poland (Klimkowski & Nagy, 2014).
To assess the impact of different reservoir and hydraulic fractures parameters on a horizontal
shale gas well production performance, sensitivity analysis (SA) was performed on a range of
parameters (variables, factors) with three sample values, called low, medium and high, for each
parameter. From the reservoir parameters matrix permeability and porosity, natural fracture
spacing and conductivity (and resulting porosity and permeability) as well as adsorbed gas con-
tent were investigated. Hydraulic treatment results were incorporated by the primary hydraulic
fracture spacing, extent and permeability. Moreover, initial water saturation in both natural and
hydraulic fractures and effect of diffusion were taken into account. The parameters used for SA
and their values are summarized in Table 1 and Table 2. The cumulative gas production was used
as the objective function of the analysis and a base case simulation model was prepared as the
reference for the results of analysis.

2. Shale gas modeling

Modeling the shale gas reservoirs differs significantly from modeling of conventional res-
ervoirs (Andrade et al., 2011). There are two main problems with organic rich shales: one is
related to complex pore system with distinct physical properties of each element of this system
and thus complex interaction between these elements, the other one results from multiple gas
storage and transport mechanisms (Arogundade & Sohrabi, 2012; Ozkan et al., 2010).
In shale reservoirs we can find three different naturally occurring porous systems (Ozkan
et al., 2011; Clarkson, 2013):
1. gas-wet organic porosity that is a mature form of kerogen,
2. inorganic, primarily water-wet, porosity system and
3. natural fractures.

Fourth pore system is generated by massive hydraulic fracturing treatment (hydraulic

fractures).
The quad-porosity model is presented in Fig. 1. The solid tanks represent four different
porosity systems and the dashed small tanks represent internal physical phenomena: A/D repre-
sents adsorption/desorption effects in organic porosity and G/W represent gas/water dissolution
/evolution effects. The valves represent the connectivity between each element of the system
(Hudson, 2011).
In numerical simulation naturally fractured reservoirs are modeled with use of one of three
main models, namely Dual Porosity (DP), Dual Permeability (DK) or Multi INteracting Con-
tinua (MINC). Each of these models has somehow limited abilities to model shale reservoirs.
For the purposes of this study solution proposed by Rubin, 2010 (Computer Modelling Group
Ltd.) and combining positive features of DK and MINC models was used. This Locally Spaced-
Logarithmically Refined-Dual Permeability model (LS-LR-DK) (Fig. 2) is based on a dual
permeability (DK) model with an additional, limited to the vicinity of the hydraulic fractures
(LS Locally Spaced) grid refinement and grid blocks are increasing in size logarithmically
away from the fractures (LR Logarithmically Refined). The result is better than the usual dual
permeability model representation of transient flow from a matrix with very low permeability

Unauthenticated
Download Date | 8/7/17 10:24 AM
990

Fig. 1. Tank representation of the quad-porosity system (Hudson, 2011)

to highly conductive hydraulic fracture. A non Darcy flow permeability based correction factor
can allow non-Darcy flow to be accurately modeled in fracture conduits modeled explicitly with
blocks as wide as 0.601 m (2 ft) (Rubin, 2010; CMG, 2013). This approach provides a very ac-
curate representation of the flow to and in the fracture (including the non-Darcy flow) at a reduced
computational power (less time-consuming simulations).

Fig. 2. LS-LR-DK fracture network in SRV (red - hydraulic fracture)

3. Simulation model description

The idealized simulation model was built based on a Cartesian grid with dimensions of
9949 9 basic blocks. Dimensions of each basic block (before refinement) are 20 20 9 m,

Unauthenticated
Download Date | 8/7/17 10:24 AM
 991

resulting in overall model dimensions of 1980 980 99 m. The horizontal well is 1400 m long
and in base case it has 15 stages of hydraulic fracturing (100 m spacing).
Gas adsorption/desorption is modeled with use of the Langmuir isotherm. The model in-
corporates non-Darcy flow in hydraulic fractures and gas diffusion in the rock matrix. A single
simulation in Sensitivity Analysis covers a period of 10 years of operation. Initial gas rate is
constrained to 200000 Sm3/day and minimum well bottom hole pressure is set to 30 bar. These
data are based on conditions used in one of the best US shale plays like Haynesville.
Natural gas is modeled as single component methane and the phenomena of change of PVT
properties in the confined porous conditions (Nagy & Siemek, 2014) is not incorporated within
the model. This is due to the fact the simulators use standard Peng-Robinson or Soave-Redlich-
Kwong equation of state to model hydrocarbons PVT behavior (PREOS in this case).
Relative permeability curves for viscous flow in shale matrix were generated with standard
correlations for 20% connate water saturation. Curves for gas and water are presented in Fig. 4.
For flow in fractures different relative permeability relationship was used, where values of kr
changes linearly from 0 to 1 with increasing phase saturation. For reference purpose initial gas
in stimulated reservoir volume of Base Case model (assumed area of SRV is presented in Fig. 3)
was assessed to be 475.8 million Sm3 (including 110.8 million Sm3 of adsorbed gas). The gas
initially in place in other cases may differ in comparison to the reference case because of dif-
ferent porosity.

TABLE 1

Parameters and their sample values used for sensitivity analysis

Parameter Unit Base case Low Medium High

Matrix permeability [mD] 0.0001 0.00001 0.0001 0.0005
Matrix porosity [-] 0.04 0.04 0.07 0.10
Natural fracture spacing, and [m] 20 5 20 30
resulting fracture porosity [-] 2 105 8 105 2 105 1.33 105
Natural fracture conductivity [mD m] 0.25 103 0.25 103 0.50 103 1.00 103
Hydraulic fracture spacing, and [m] 100 100 200 350
number of fracturing stages [-] 15 15 8 5
Hydraulic fracture extent (one wing), xf [m] 150 50 90 150
Hydraulic fracture permeability, kf [D] 5 1 5 20
Adsorbed gas [Sm3/tonne] 1 0 1 5
Diffusion [cm2/s] 0.0006 0 0.0003 0.0006
Fracture water saturation [-] 0.55 0.20 0.35 0.55

TABLE 2

Natural fracture permeability (nD) depending on natural fracture conductivity and spacing

Spacing [m]
5 20 30
Conductivity [mD m]
3
0.25 10 50.0 12.5 8.3
0.50 103 100.0 25.0 16.7
1.00 103 200.0 50.0 33.3

Unauthenticated
Download Date | 8/7/17 10:24 AM
992

Fig. 3. Assumed area of SRV for initial gas in place calculation

Fig. 4. Relative permeability curves used for viscous flow in matrix

4. Results of Sensitivity Analysis

The input for SA consists of 10 parameters with 3 sample values for each, resulting with
parameter space (the number of all possible experiments) equal 310 = 59049. Obviously it is not
possible to perform such an analysis and thus it is necessary to select some reasonably limited
set of experiments. According to statistical experimental design theory, to efficiently explore the
parameter space, the design (the set of experiments) selected should be a representative subset
of all possible experiments. The Latin hypercube design approach was applied to reduce the

Unauthenticated
Download Date | 8/7/17 10:24 AM
 993

number of experiments and meet above mentioned requirement. Finally 81 simulations were
performed within the SA. The illustrative summary of these runs in comparison with base case
simulation results (thick red lines) is presented in Fig. 5. At first glance, the base case ranks in
the middle range of SA results.
Based on results of all simulation runs linear and reduced quadratic proxy models were cre-
ated and then used to determine main (linear) and interaction/quadratic (nonlinear) effect estimates
of particular parameters. The effect estimate show how changing the parameters value change
the models response (objective function). The linear proxy model is described by the equation:
y a0 a1 x1 a2 x2 ... ak xk

where y is the models response (objective function), a0, a1, ..., ak are the parameters of the proxy
model, and x0, x1, ..., xk are the input variable parameters.
In addition to linear effects (main effects), parameter interaction effects (cross terms xi xj)
and quadratic effects (xj2 ) can be extracted from second degree (quadratic) polynomial model:
k k k
y a0 a j x j a jj x2j aij xi x j
j 1 j 1 i j j 2

Fig. 5. Gas production rate (top) and cumulative gas production (bottom) for all simulation runs
of sensitivity analysis (thick red line represents base case simulation)

Unauthenticated
Download Date | 8/7/17 10:24 AM
994

At this point it is worth recalling that proxy model is just an approximation of the simulation
model and actual change (increase or decrease) of the objective function due to the change of
one parameter varies depending on the combinations of sample values of the other parameters.
Key results of SA are presented in the form of tornado plots. Parameters are ordered from
having the greatest effect on the objective function to having the least effect. The Maximum
and Minimum bars represent the maximum and minimum objective function (cumulative gas)
value among all the training jobs. Between them base case cumulative production is presented
for comparison.
Linear model effects on cumulative gas production in 10 years period are shown in Fig. 6.
Minimum and maximum cumulative gas production are 15.5 and 427.0 million Sm3 respectively,
while the base case production reaches 188.1 million Sm3. Parameters of the adequate simulation
runs are summarized in Table 3. Analyzing the results one should remember that the worst and
the best runs of the set of experiments are not necessarily the worst and the best configurations
of sample values possible. They are such in this particular Latin Hypercube design. Also it should
be noted that due to the fact that shale rock porosity is one of the parameters being investigated
in SA, initial gas in place amount vary between the runs. Thus, recovery factor in, for example,
maximum or minimum production case cannot be calculated based on the reference case original
gas in place. However, the aim of the analysis is, based on statistically representative subset of all
possible experiments, to figure out shale gas simulation key factors, and not to focus on recovery
factor of each experiment (simulation run).

TABLE 3

Juxtaposition of parameters of minimum and maximum simulation runs

Parameter Unit Minimum case Maximum case

Matrix permeability [mD] 10 500
Matrix porosity [-] 0.04 0.10
Natural fracture spacing, and [m] 30 5
resulting fracture porosity [-] 1.33 105 8 105
Natural fracture conductivity [mD m] 0.25 103 1.00 103
Hydraulic fracture spacing, and [m] 350 100
number of fracturing stages [-] 5 15
Hydraulic fracture extent (one wing), xf [m] 90 90
Hydraulic fracture permeability, kf [D] 1 5
Adsorbed gas [Sm3/tonne] 1 1
Diffusion coeficient [cm2/s] 0.0006 0.0003
Fracture water saturation [-] 0.35 0.35
SRV initial gas in place [106 Sm3] 476.6 1018.4
Cumulative gas [106 Sm3] 15.5 427,0

According to Fig. 6 rock matrix permeability has the biggest impact on overall well perfor-
mance. This means that increase of matrix permeability from 10 to 500 nD causes expected increase
of cumulative gas production of 143.2 million Sm3. According to linear proxy model the second
and the third highest impact on the objective function belongs to hydraulic fracture spacing and
extent, which in total is good information, since both these parameters are (more or less) under
control during fracturing treatment. Next most important parameter is natural fracture spacing,

Unauthenticated
Download Date | 8/7/17 10:24 AM
 995

Fig. 6. Linear model effects on cumulative gas production

and what is interesting from shale modeling point of view, conductivity of natural fractures is not
of very high importance. Density of natural fracture network has a much greater importance than
its conductivity. Based on linear proxy model the fifth biggest impact on production belongs to
matrix porosity: increase from porosity of 0.04 up to 0.10 causes expected increase of cumulative
gas production of 52.3 million Sm3. Whereas the effect of diffusion, desorption and initial water
saturation of fractures is much smaller than properties of shale rock.
It should also be pointed that, although the extent and density of hydraulic fractures in
stimulated reservoir volume are key performance indicators, hydraulic fracture permeability has
the least impact on the well performance.
According to Fig. 7 three most important effects are the linear effects of matrix permeability
and hydraulic fractures spacing and extent. Next is the quadratic effect of matrix permeability
and interaction effects of hydraulic fracture spacing with extent and matrix permeability.

Unauthenticated
Download Date | 8/7/17 10:24 AM
996

Fig. 7. Reduced quadratic model effects on cumulative gas production

5. Extension of the Sensitivity Analysis

5.1. Focus on key factors

To extend the results of sensitivity analysis base case model was modified in the one-
parameter-at-time manner to look more closely at the impact of the parameters indicated by the
SA. In subsequent simulation runs effects of hydraulic fracture spacing and permeability, matrix

Unauthenticated
Download Date | 8/7/17 10:24 AM
 997

permeability, natural fractures parameters and adsorbed gas content were analyzed. Additionally
production period was extended up to 20 years. Results of these simulations are presented in
Fig. 8 to Fig. 13.

Fig. 8. Matrix permeability impact on the cumulative gas production (recovery of 260 mln Sm3 during
20 years of production assuming matrix permeability 100 nD)

Fig. 9. Hydraulic fracture spacing impact on the cumulative gas production for 100, 200 and 350 m
distance between fractures

Unauthenticated
Download Date | 8/7/17 10:24 AM
998

Fig. 10. 3D cross-section along the well: left side pressure, right side adsorbed gas after 20 years
of production; top to bottom hydraulic fracture spacing 100, 200 and 350 m

Unauthenticated
Download Date | 8/7/17 10:24 AM
 999

Fig. 11. Natural fracture spacing impact on the cumulative gas production for I and J direction spacing
of 5, 20 and 30 m and basic grid block dimension of 2020 m

Fig. 12. Adsorbed gas impact on the cumulative gas production for 0, 1 and 5 Sm3/tone

Unauthenticated
Download Date | 8/7/17 10:24 AM
1000

Fig. 13. Hydraulic fracture conductivity impact on the cumulative gas production 2.5, 12.5 i 50 mD m
(or permability of fractures 1, 5, 20 D)

5.2. Impact of results of fracturing treatment and SRV

modeling approach
Performed sensitivity analysis confirmed that properly designed and executed fracking is
crucial for shale gas production performance. Consequently appropriate modeling approach is
required to account properly for the fracking effects. Beside standard planar hydraulic fracture
a stimulated reservoir volume (SRV) approach is used, based on a tartan grid (Fig. 14).
Hereafter effects of fracturing treatment quality, including so called stress shadowing,
when stress field changed by one fracturing stage can affect another one, are presented. This is
represented by reduced extent of some fractures (8 full and 7 limited-extent fractures) or com-
pletely not effective treatment in some stages (8 full-extent fractures) compared to 15 full-extent
fractures case (Fig. 15). In practice, the efficiency of the fracturing oscillates in the range of 60-
70%. Also impact of different representations of the SRV is presented, comparing set of single
planar fractures to set of one block wide (20 m) tartan SRV and 3 blocks wide (60 m) tartan
SRV (Fig. 16). These runs were performed with higher adsorbed gas content than the above.

Unauthenticated
Download Date | 8/7/17 10:24 AM
 1001

Fig. 14 Tartan grid SRV: set of 3-blocks-wide fracturing stages

Fig. 15 Impact of quality of stimulation process on production performance.

Different rate and cumulative production as result of density (spacing) and extent of hydraulic fractures
(15 normal fractures, 8 normal + 7 poor quality fractures and 8 normal fractures)

Unauthenticated
Download Date | 8/7/17 10:24 AM
1002

Fig. 16 Impact of the SRV modeling method (15 individual hydraulic fractures, SRV zone composed using 15
1-block-wide tartan SRVs and 15 3-blocks-wide SRVs)

6. Conclusions
1. Hydraulic fracture spacing and extent has strong influence on well performance. However,
fracture permeability is not of very high importance. There is an evident increase in production
between hydraulic fracture permeability of 1 D and 5 D, especially in the initial period of pro-
duction but increasing fracture permeability up to 20 D does not increase cumulative production
much. Moreover, in later period there is no difference in production rate. Summing up, as the
contrast between the rock matrix and fracture is so significant it is not necessary to have very
high conductive hydraulic fractures. The focus should be on obtaining dense and wide stimulated
reservoir volumes (SRV) with complex fracture network.
2. Relatively high permeable shales combined with long and dense hydraulic fractures
(effective SRV) can yield improved performance, especially in later phase of production and in
shales with low natural fracture permeability. The maximum matrix permeability in this work
has been selected as 500 nD, which may be higher than expected in normal situations and is
representative only to the best sweet spots.
3. It is necessary to properly determine natural fracture distribution. Relatively dense natural
fracture network increases production performance in early phase of production (Fig. 11). Also,
more advanced relative permeability model for flow in fractures may be of interest. Conductivity
of natural fractures is not of very high importance.
4. High adsorbed gas content can have important impact on production performance in late
time production (Fig. 12).

Unauthenticated
Download Date | 8/7/17 10:24 AM
 1003

Acknowledgements

The authors are grateful to the National Centre for Research & Development for providing
research fund through Blue Gas Program (projects IRES) and statutory research program
(11.11.190.01/2014) performed at the AGH University of Science and Technology (Faculty of
Drilling, Oil and Gas).

References

Ambrose R.J., Hartman R.C., Diaz-Campos M., Akkutlu I.Y., Sondergeld C.H., 2010. New Pore-scale Considerations
for Shale Gas in Place Calculations. Paper SPE 131772-MS. doi 10.2118/131772-MS.
Andrade J., Civan F., Devegowda D., Sigal R.F., 2011. Design and examination of requirements for a rigorous shale-
-gas reservoir simulator compared to current shale-gas simulator. Society of Petroleum Engineers Paper 144401.
Arogundade O., Sohrabi M., 2012. A Review of Recent Developments and Challenges in Shale Gas Recovery. SPE
paper 160869.
Clarkson C.R., 2013. Production data analysis of unconventional gas wells: Review of theory and best practices. Inter-
national Journal of Coal Geology, 109-110, 101-146.
CMG, 2013. GEM Users guide. Compositional & unconventional reservoir simulator.
EIA, 2013. Technically Recoverable Shale Oil and Shale Gas Resources: An Assessment of 137 Shale Formations in 41
Countries Outside the United States. June 10, 2013
Hudson J., Civan F., Michel-Villazon G., Devagowda D., Sigal R., 2012. Modeling Multiple-Porosity transport in gas-
-bearing shale formations. SPE 153535.
IEA, 2012. World Energy Outlook: Golden Rules in the Golden Age of Natural Gas. Report IEA.
IEA, 2011. World Energy Outlook 2011: Golden Age of Natural Gas?. Report IEA.
Kaliski M., Nagy S., Siemek J., Sikora A., Szurlej A., 2012. Unconventional natural gas-USA, the European Union,
Poland. Archiwum Energetyki 42, 109-122 (In Polish).
Klimkowski ., Nagy S., 2014. Symulacja numeryczna wydobycia gazu ze stymulowanej objtoci zoa upkowego.
Midzynarodowa Konferencja Naukowo-Techniczna Geopetrol 2014, Poszukiwania i eksploatacja z ropy naftowej
i gazu ziemnego nowe technologie, nowe wyzwania, September 15-18, 2014 Zakopane, p. 777-781 (In Polish).
MIT, 2011. Future of Natural Gas, An Interdisciplinary MIT Study. Massachusetts Inst. of Tech., Boston.
Nagy S., Siemek J., 2011. Shale Gas in Europe: the state of the technology challenges opportunities. Arch. Min. Sci.,
Vol. 56, No 4.
Nagy S., Siemek J., 2014. Confinement Phase Envelope of Gas Condensate Systems in Shale Rocks. Arch. Min. Sci.,
Vol. 59, No 4, p. 1005-1022.
Nagy S., Siemek J., Wisniowski R., 2013. Unconventional Energy Resources: New Challenges in Research and Education.
Proc. of ICEE ICIT 2013 Conference, Captown, SA, Dec. 2013.
Nagy S., Rychlicki S., Siemek J., 2009. Stan obecny i ewolucja stosunkw gazowych Rosji z Uni Europejsk i Polsk.
Polityka Energetyczna 12, 393-422 (In Polish).
Ozkan E., Brown M., Raghavan R., Kazemi H., 2011. Comparison of fractured horizontal-well performance in tight
sand and shale reservoirs. Society of Petroleum Engineers Reservoir Evaluation & Engineering 14 (2), 248-259.
Ozkan E., Raghavan R., Apaydin O.G., 2010. Modeling of fluid transfer from shale matrix to fracture network. Society
of Petroleum Engineers Paper 134830.
PGI, 2012. Assessment of shale gas and shale oil resources of the Lower Paleozoic Baltic-Podlasie-Lublin Basin in
Poland. Report, Warsaw 2012.
Poprawa P., 2010. Poszukiwanie z gazu ziemnego w upkach (shale gas) w Polsce. Wiadomoci Naftowe i Gazownicze
2/2010 (In Polish).

Unauthenticated
Download Date | 8/7/17 10:24 AM
1004

Rubin B., 2010. Accurate simulation of non-Darcy flow in stimulated fractured shale reservoirs. SPE 132093.
Siemek J., Nagy S., 2012. Energy carriers use in the world: natural gas conventional and unconventional gas resources.
Arch. Min. Sci., Vol. 57, No 2, p. 283-312.
Siemek J., Nagy S., Siemek P., 2013. Challenges for sustainable development: the case of shale gas exploitation in
Poland. Problems of Sustainable Development, vol. 8, no. 1, p. 91-104.
Siemek J., Rychlicki S., Kolenda Z., Stryczek S., Nagy S., 2009. Gaz ziemny i ropa naftowa Polska i Unia Europejska,
potrzeby, wasne wydobycie i dostawy. Zarys stanu i perspektywy energetyki polskiej. Studium AGH, AGH Krakw
(In Polish).

Received: 22 April 2014

Unauthenticated
Download Date | 8/7/17 10:24 AM